Ground Penetrating Radar (GPR) data processing is a systematic project. Firstly, the dissertation analyses the fault of existent software and introduces the development of GPR data processing history at home and abroad. Then this dissertation summarizes and presents the basic principles of GPR and discusses the key factors which influence its resolution and the depth that it can penetrate. The dissertation points out the possible origin of noise that affects the quality of GPR data with combination of actual data and forward modeling. Based on the former analysis, the author designs GPR Processing Software under the environment of Matlab and Visual C++. Data format analysis and converting are completed by C++ and the main program is made in Matlab. The main program includes four parts: File Operation, View Operation, Basic Edit and Filtering. File operation includes reading and saving files, editing file headers and format convert. The View menu offers data visualization options, which includes image (color-coded) displays with several color maps for better discrimination of the details, wiggle-trace displays and variable-area displays and gain manipulation. It is also possible to plot individual traces and trace spectra, as well as the instantaneous attributes (instantaneous amplitude, instantaneous phase and instantaneous frequency) of the input data and processed data. Basic Edit includes adjustment of signal time-zero, basic trace editing (trimming and extraction), resampling and marker interpolation routines is also provided to data at equal-distance spacing. Filtering provides a suite of data smoothing facilities. These include: zero-phase FIR filtering of the frequency content and of the wave number content, interactive F-K filter design and implementation, Mean and median spatial filtering in one and two dimensions and removal of a sliding window mean (background) trace. The software works in an interactive way. It is friendly to use and the results are returned to customers using graphic scheme. Thus that it is convenient to monitor and analysis of changes caused by operation. At last, the dissertation gives an example to prove it efficiently by using processing method provided by the software. |